Giant fornix syndrome: a recently described cause of chronic purulent conjunctivitis and severe ocular surface inflammation, with a new diagnostic sign on CT

Sir,

Case report

A 75-year-old lady initially presented with a red, sticky right eye for several weeks with copious yellow discharge. She had no prior ophthalmic history. Her medical history included rheumatic fever resulting in mixed mitral valve disease and pulmonary hypertension.

At presentation, she was noted to have a small peripheral corneal epithelial defect with mild infiltrate and was diagnosed as having blepharoconjunctivitis and marginal keratitis and treated with guttae chloramphenicol and topical steroid (maxidex). Lid hygiene was also advised. A conjunctival bacterial swab grew Staphylococcus aureus, sensitive to, among others, chloramphenicol.

One week later, the epithelial defect had improved but there had been little improvement in the discharge and she was kept under outpatient review as she suffered recurrent episodes of purulent discharge. Repeated cultures grew S. aureus with identical sensitivities. The absence of a mucocele was noted and syringing and probing revealed patent flow through the nasolacrimal duct. Her left eye remained unaffected.

Fourteen months after her initial presentation, she returned acutely complaining of increased discharge and unusually severe pain and photophobia. Her right eyelids were matted together and upon cleaning there was found to be a large volume of yellow discharge emanating from beneath the upper lid, which required irrigation and furthermore, hampered the remaining examination. The eye was injected and there were signs of early corneal thinning nasally. A diffuse epitheliopathy and areas of limbal vascularisation were also present. Again, there was no mucocele. An urgent Gram stain of a conjunctival swab showed Gram-positive cocci (and later grew sensitive S. aureus) and on microbiological advice she was recommenced on guttae chloramphenicol.

On review 3 days later, there had been an improvement in the discharge, but the right eye remained very injected and a corneal epithelial/anterior stromal defect was evident, with no infiltrate. Despite intensive topical and oral antibiotic treatment, the nasal corneal thinning progressed with descemetocele formation (Figure 1) and signs of sclerokeratitis. Investigations at that stage revealed an ESR 56 and CRP 13.3, Rheumatoid Factor 87.9 (<20) and a positive ANA 1:40. ANCA were negative.

Figure 1

Severe ocular surface inflammation with anterior stromal defect and limbal descemetocele.

She was commenced on 500 mg oral methylprednisolone for 3 days, followed by a reducing dose of daily oral prednisolone. Medical and rheumatological review reported no signs of active rheumatoid arthritis or evidence for a systemic vasculitis. Her right eye settled over the following week with no further progression in corneal thinning and the epithelial defect healed.

In view of the intermittent recurrence of discharge a CT scan of her orbits and paranasal sinuses was performed. There was no sign of foreign body or frontal sinusitis. However, a pocket of air was noted deep in the upper fornix of the right eye (Figure 2). On subsequent examination, our patient was noted to have age-related dehiscence of the levator aponeurosis (Figure 3) and a diagnosis of Giant fornix syndrome was made. Her eye remains quiet with occasional upper fornix irrigation and on maintenance topical steroid and antibiotic therapy.

Figure 2

CT scan of the orbits demonstrating air in the right superior conjunctival fornix.

Figure 3

Age-related dehiscence of the levator aponeurosis and orbital fat atrophy.

Discussion

Rose¹ published a case series of 12 patients (10 females) aged between 77 and 93 presenting to Moorfields Eye Hospital with chronic relapsing purulent conjunctivitis affecting one eye and lasting between 8 and 48 months. Three had undergone successful DCR before the diagnosis of Giant fornix syndrome was made. Nine had corneal vascularisation and scarring before referral. Five had suffered spontaneous perforation or thinning. As in this case, all had deep upper fornices and associated changes of age-related dehiscence of the levator aponeurosis and universally, S. aureus was the inhabitant.

He postulated that gradual deepening of the upper fornix owing to age-related disinsertion of the levator aponeurosis allowed for the accumulation of a bacteria-laden protein coagulum within a capacious upper fornix. Persistent reinoculation of the tear lake by low-grade bacteria from the fornix, severe conjunctivitis from ‘toxic’ bacteria and the development of a pseudomembrane, may enhance the coagulum and also exacerbate the ptosis and deepening of the upper fornix. This environment leads to severe ocular surface inflammation.

This case highlights the need to consider a deep upper fornix, present in many of our patients with age-related levator disinsertion and orbital fat atrophy, as the source of a recurrent (often unilateral) purulent discharge. In addition, we suggest that CT imaging is beneficial in excluding frontal sinus fistulae and foreign bodies in such cases, and the presence of air in the deep upper fornix of this patient on CT scan is a new sign of GFS, not described previously.